As a lab technician at the University of Vermont in Burlington, Walter DeNino had a gut feeling that something was amiss when his boss repeated a statistical analysis that DeNino had done and arrived at a different conclusion. DeNino followed his hunch and spent his evenings reconstructing the data set from patient medical records. He found that the original data did not match his supervisor's figures.
DeNino had uncovered one of the most serious cases of scientific misconduct reported in recent years. His boss, obesity expert Eric Poehlman, had committed scientific fraud for more than 12 years in numerous publications and grant proposals (see 'Obesity expert owns up to million-dollar crime'). Now debarred from receiving federal research funding for life, Poehlman must repay $180,000 and is one of only two researchers ever charged in a US criminal court for misconduct.
"I felt that his behaviour had to be exposed and that he should be removed from science," says DeNino, now a researcher in another obesity lab group at Weill Medical College, Cornell University in New York. "But at the onset, I didn't imagine it would take five years and that criminal charges would be filed."
After an arduous and sometimes ugly investigation process, DeNino says the experience opened new research opportunities for him and spurred his desire to become a physician-researcher. But he admits he was lucky. As a whistleblower, he was very cautious in making an allegation and he was well-protected by the policies and actions of the University of Vermont.
Known cases of misconduct are rare, involving one or two scientists in 100. Chances are, however, that most scientists at some time will come across a lab member's notebook, a collaborator's work, a manuscript, grant proposal or thesis that seems suspect. Knowing the proper way to approach and investigate will help you handle such a sticky situation.
Lessons can be learned from the cases of three junior scientists who found misconduct by senior researchers, did or did not come forward, and the consequences of their actions. Misconduct by a supervisor puts the junior scientist in a very vulnerable position. But with the proper tools and advice, a graduate student, postdoc or junior faculty member should be able to make an allegation without risking her or his career.
Corrupt or just sloppy?
It is vital to know what constitutes misconduct. The US Public Health Service defines it as "fabrication, falsification or plagiarism in proposing, performing or reviewing research, or in reporting research results" (see 'Office of Research Integrity -Policies - Regulations'. It does not include honest mistakes or differences in interpretation. It must be shown to have been committed intentionally or recklessly and be a significant departure from accepted practices in a field. This 'FFP' definition is widely accepted by universities and other government agencies.
Some unethical or poor research methods fall short of being FFP. They include keeping inadequate records, putting speculation forth as fact, using confidential information and dropping data points from statistical analyses. These are much more common than FFP, with close to a third of scientists engaging in such questionable practices (see 'Scientists behaving badly').
"We as a community need to work on research integrity a little bit more," says Julio Turrens, associate dean and biochemist at the University of South Alabama in Mobile. He teaches courses on research integrity in the university's school of medicine.
Although misconduct cases are few, their effects on the researchers involved, the field and ultimately on society can be far-reaching, says Turrens. He gives students examples of fraudulent studies that changed environmental regulations or had breast-cancer patients clamouring for an ineffective treatment. He warns students that if they make an allegation of misconduct to him, as a dean he is bound to investigate it. Most institutions have a protocol for investigating scientific fraud, and Turrens advises scientists to consult two or three sources of confidential advice before passing "the point of no return".
DeNino was successful because he knew where that point was in his university's investigation policies. He set up an anonymous e-mail account so that he could get advice and information from university offices before stepping forward formally.
Turrens suggests doing your homework to find the university office that handles investigations — often called the 'office of research compliance' or 'research integrity' — and figuring out how to report something directly to that office. "Often that puts the allegation out of the loop of the department or school, then there is no allegiance between the person you are talking to and the person you are complaining about," he says.
DeNino got trusted colleagues to double-check his suspicions, which research-integrity experts strongly recommend. Margaret Dale, dean for faculty and research integrity at Harvard Medical School, advises finding someone trustworthy to test that you really understand what's happening. "Ask in a non-threatening way, 'I see this happening, what do you think?'," she says. Then if you do proceed "there's another person who knows what's going on and safety in numbers". (See box)
Kent*, a postdoc in a US West Coast institution, says his allegation against his former adviser might have had a better outcome had he found senior scientists to back him. "It is very difficult to find faculty members to take your side in public," he says. Although some supported his findings, they were unwilling to confront his adviser. Kent's situation highlights the vulnerability of postdocs who come into an institution with a one-to-one relationship with an adviser. It easily becomes a case of one's word against the other's.
Kent discovered his adviser had taken parts of a grant proposal Kent had written based on his graduate work and used them, verbatim, for another postdoc's fellowship application, giving no credit to Kent's previous lab and work. Kent filed a formal grievance with the vice-chancellor for research. But because the original proposal was submitted under his adviser's name, and because Kent had no proof of intent to steal credit, the grievance resulted in a rather unsatisfying letter of apology — and, Kent believes, retaliation.
Julio Turrens: "We need to work on research integrity.""In this situation, you are pretty powerless," Kent says. "Be very careful about writing a grant application submitted under someone else's name. The investigator has too much power in that situation." He advises getting a written contract that spells out your expected contribution to research and the time frame you are expected to work in.
Integrity officers stress that junior scientists must find senior scientists or officials who will support them if they bring forward an allegation. Dale says postdocs especially should find out what resources are available to them — postdoc office, ombudsman office, and so on — before moving ahead with a complaint.
Dale notes that, with most allegations arising within a lab group, it is nearly impossible to keep the accuser's identity confidential from the accused. Junior scientists must understand the risks they will face if they become known as a whistleblower and the consequences their actions may have, such as shutting down a laboratory.
Stakes too high
Emma* weighed up the risks against some dubious practices in her graduate thesis laboratory at a private research institute. Although her adviser verbally presented data at meetings that she knew to be false, she did not speak out. She and lab colleagues discussed the matter, but felt the stakes were too high, as some would lose jobs or prospects if the lab was disgraced.
Such conflicts are normal, says Chris Pascal, director of the Office of Research Integrity, which monitors investigations of misconduct in research supported by US Public Health Service funds. "Unless the misconduct is so serious that a human life is at stake, I would never say that someone has to come forward," Pascal says. "That is a really personal decision left to the individual."
Dale comments that once an FFP misstep reaches publication stage, whistleblowers who are co-authors must also step forward or be found complicit later.
"It's an honour system," says Mortimer Litt, associate dean and senior scientific investigator at Harvard Medical School. "We do not have thought police monitoring every lab notebook every day. We depend on whistleblowers."
Stepping into that role is not easy. DeNino found that out when he was counter-accused of planting the falsified data. "It's not something I would suggest doing if you don't have a strong character," he says. "You can never predict what will happen."
It's a roll of the dice — but so is remaining quiet, says Turrens. "You may choose to ignore it, but look at it from a selfish standpoint. If the person is ever caught, your career will have a spot on it."
CSI: Harvard
When an allegation of scientific misconduct is made at Harvard Medical School, Mortimer Litt, senior scientific investigator, springs into action. His investigation may not be drama-filled like forensics television shows, but it does involve reconstructing what happened.
"In academia, we value nothing more than accurate reporting. We pursue it to the bitter end," he says. He and colleagues go to the lab to secure original data and lab notebooks so they can't be altered. Litt does a preliminary investigation and reports his factual findings to a panel of three senior professors.
This panel determines if there is enough "presumptive evidence of misconduct" to send it to the larger standing committee for investigations. That committee delivers a decision and recommendations for sanctions, if any, to the accused's dean and to the funding agency, if the case involved grants.
Litt says the process is not a trial, but a search for the truth. The accused gets copies of all notes and testimony; nothing is sprung on them Perry Mason-style. The process can take from a couple of months to six years. Sometimes lab work continues, but sometimes the accused is banned from the lab until the matter is resolved.
Litt enjoys using his intellect to solve such mysteries — comparing duplicated images or forged signatures. "I always get a thrill when I make connections," he says. "Then, my next thought is that this is a human tragedy."
K.P.
